fix(skills): prevent premature thread resolution on self-authored PRs#1585
Open
worktrunk-bot wants to merge 1 commit intomainfrom
Open
fix(skills): prevent premature thread resolution on self-authored PRs#1585worktrunk-bot wants to merge 1 commit intomainfrom
worktrunk-bot wants to merge 1 commit intomainfrom
Conversation
The review bot incorrectly resolved its own technically-correct suggestion on PR #1583. It had correctly identified that linked worktrees can be on different mount points (EXDEV on rename), but in a later incremental review run it accepted the author's less accurate doc comment claiming ".git/ is always on the same filesystem" and resolved the thread. Add guidance to Step 6 (thread resolution) requiring: - Verification that the substance was addressed, not just reworded - Extra skepticism on self-authored PRs where the bot is both author and reviewer, creating a bias toward accepting its own code's claims Evidence: run 23182191893, PR #1583, session e2cae4a2 Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Evidence
Run 23182191893 reviewing PR #1583 (
fix/issue-1572-trash-dir):The bot's earlier review correctly identified that linked worktrees can live on different mount points, so
rename()from a linked worktree path to.git/wt-trash/could fail with EXDEV. It suggested a doc comment:In run 23182191893 (incremental review), the bot encountered the author's version which claims "The
.git/directory is always on the same filesystem as the worktree (git requires this)" and reasoned "The author's version is actually more accurate." It then resolved the thread.The original suggestion was correct — git does NOT require
.git/and linked worktrees to share a filesystem. The.gitentry in a linked worktree is a file pointing to<git-common-dir>/worktrees/<name>, and that common dir can be on any mount point. The code handles this correctly (fallback to legacy removal on rename failure), but the doc comment is misleading.Root cause: When the bot authored the PR and then reviews it, there's a bias toward trusting the code's own claims rather than independently verifying them.
Other observations (no action needed)
running-in-ci, unlike all other review sessions. Intermittent model behavior, not fixable via skill changes.Sessions analyzed
🤖 Generated with Claude Code